Scientific method seemed to many in the nineteenth century to be somehow the essence of science. Learning science was hard: lots of facts, difficult language, symbols and concepts, and often all of very remote relevance to ordinary life. Only some people were ever going to need to master it; but perhaps others could pick up the scientists’ method of finding truth and apply it elsewhere, so that their conclusions were not dominated by emotion, deference to authority, or mere habit. The motto of the Royal Society was, and is, ‘Nullius in Verba’, take nobody’s word for it. Careful testing was required, though it could be ambiguous. It was said that a spider could not escape from a circle of unicorn’s horn, and the early Fellows duly tried the experiment. The spider got away, which showed either that the claim was false, or that their unicorn’s horn was not genuine. That kind of difficulty attends all experiment, though in practice scientists try to control the circumstances so that only one option is really at all likely, and the result therefore convincing. Francis Bacon’s hope was that we could rid ourselves of all our ‘idols’ or preconceptions, but that seems optimistic; science does rest upon some metaphysical assumptions to do with uniformity, which is why psychical research is so problematic – telepathy, even when no trickery is apparent, seems to have good and bad days.